Referring now to the prior art in this field of engineering with reference to a disc memory by way of example, the disc memory of a conventional type is of the construction in which magnetic paint prepared by mixing fine powder of .gamma.-Fe.sub.2 O.sub.3 with an epoxy resin or other binding agents has been applied to a metallic aluminum disc and then hardened by being heated to a hardening temperature of the binding agent so as to fix the .gamma.-Fe.sub.2 O.sub.3 to the aluminum disc. This conventional type disc memory thus obtained has the following problematic points with respect to the article or with respect to the method of manufacture.
1. With respect to the article:
(A) A ferrite film depends upon the hardness of the binding agent used for its surface hardness and, generally speaking, it is liable to damage by the contact friction of a magnetic head for writing or reading. This may result in the film not only reducing the capacity of the magnetic memory but often may cause the detachment of the film from the magnetic memory, losing the memory capacity of the memory structure entirely. Particularly, when it becomes necessary to bring the magnetic head as near the memory as possible in order to increase memory density, the damage which thus may be caused by the head to the magnetic medium is not to be ignored.
(B) Because fine powder of .gamma.-Fe.sub.2 O.sub.3 is dispersedly fixed in the bonding agent, care must be taken to disperse the powder very uniformly or otherwise there is likely to be produced discrepancy in memory characteristics. In this case, caution must be used in the dispersibility and also uniformity of grain diameter and grain shape of fine powder of .gamma.-Fe.sub.2 O.sub.3. Such thorough uniformalization is a very difficult task from a viewpoint of mass production, and in this sense the intended uniformalization ultimately results in increased cost of production. In any event, uniformity of quality in the end product is difficult to obtain.
(C) The magnetic memory structure according to the binding agent application and drying method is limited to film thicknesses such that the thinnest possible film obtainable should not be less than about 5.mu. . As film thickness is increased, memory density is reduced as is well known (thickness loss).
2. In respect to a method of manufacture:
(A) Grinding was necessary to smooth the surface of the ferrite film after the binding agent was hardened, and this grinding required much skill and lacked productivity. In addition, a thin film was very difficult to obtain as described above in Item (c).
(b) As described in Item 1. (b), it was extremely difficult to uniformalize the grain diameter, size and dispersibility of .gamma.-Fe.sub.2 O.sub.3 powder, and accordingly in order to obtain a uniform product, productivity was reduced and production cost had to be inevitably increased.
There have been numerous unsuccessful attempts to solve some of these problems, such as that shown in "Chemical Deposition and the Formation of Mixed Ferrite Films" by William L. Wade, Jr., et al, December 1964, AD 611-774; "Formation and Deposition of Ferrite Films" by William L. Wade, Jr., et al, February 1962, AD 282-515; and the U.S. patent applications showing the same Ser. No. 26,579 filed Apr. 8, 1970, and Ser. No. 95,692 filed Dec. 7, 1970, by Bernard Jacobs, et al.
The present inventors faced with such problematic points have developed, after various research and tests, an entirely new magnetic structure for discs, etc., which overcomes the disadvantages of the kinds described and a method of manufacturing the same.